Trending: Interview: ‘Bad Blood’ author John Carreyrou on Elizabeth Holmes and the Theranos scandal, a ‘what not to do’ guide for startups

Shawn DuBravac speaks at CES. (GeekWire Photo / Monica Nickelsburg)

LAS VEGAS —Alexa, will this be the CES show where your dominance is established in voice computing?

Shawn DuBravac, the chief economist for the Consumer Technology Association, thinks it very well could be.

Speaking to journalists Tuesday at the giant conference here, DuBravac predicted a huge surge in announcements around Amazon.com’s blockbuster digital assistant.

You can already ask Alexa to read your Fitbit results, or request an Uber. But at the 2017 CES, expect even more.

“You will see that list grow,” said DuBravac. “I would not be surprised if that list grew 50 percent over the next four days. There’s about … 1,500 different applications that you can do with Amazon Alexa. I would not be surprised to see 700 launch over the next four days.”

Those announcements have already started, with PC maker Lenovo and Dish Network already announcing Alexa-powered and supported devices at the show.

Lenovo debuted its own voice-activated speaker, the Smart Assistant, which runs on Alexa. Dish said its customers can now pair a Hopper DVR with an Alexa-powered speaker, like the Amazon Echo and control it using their voices.

DuBravac said that “voice is going to be the glue” to bind us to the next wave of computing, especially around advancements in the so-called smart home. You’ll see it integrated with household appliances, like televisions and dishwashers.

He estimates that about five million voice-activated assistants have been sold to date, with another five million likely to sell in 2017.

Shawn DuBravac speaks to reporters at CES. (GeekWire Photo / John Cook)

“Look for wide deployment of voice, but also look for progression of voice, and how voice is getting better and how it is being used in nuanced ways,” said DuBravac, who noted that speech recognition really got rolling at Microsoft in 1994 when the company formed a team around the technology.

In 1994, he noted that speech recognition “did not work at all.”

But in the last three years, the word-error rate in speech recognition has moved from about 25 percent to five or six percent, which is similar to human speech.

“We have seen more progress in the last 30 months than we saw in the first 30 years,” said

GeekWire reporter Monica Nickelsburg contributed to this story.

Like what you're reading? Subscribe to GeekWire's free newsletters to catch every headline

Comments

Job Listings on GeekWork

Find more jobs on GeekWork. Employers, post a job here.